Clinicians and veterinarians often select antibiotic therapies for infections on the basis of laboratory test results. The laboratory tests, known as antimicrobial or antibiotic susceptibility tests, determine the inhibitory activity of antibiotics against the microorganisms that cause infections. If the antibiotic susceptibility test indicates that an antibiotic is sufficiently potent to treat an infection, the microorganism causing the infection is reported to be “susceptible” to the antibiotic. If the test indicates a lack of sufficient antimicrobial potency for successful therapy, the microorganism is reported as “resistant” to the antibiotic. In some tests other categories of susceptibility may be reported, e.g. “moderate susceptibility” or “intermediate susceptibility.”
A problem with currently available antimicrobial susceptibility tests is their failure to reliably predict the outcome of therapy. Sometimes an antibiotic will fail to cure an infection even though the microorganism is susceptible to the antibiotic in the laboratory test. That is, the current routine laboratory tests can be misleading and give an over-optimistic impression of the therapeutic potential of antibiotics. These tests can therefore cause patients to be given ineffective treatments. In serious infections, this inadequacy of current laboratory tests can have fatal consequences.
There are many reasons for failures of antibiotic therapies that were initiated on the basis of antibiotic susceptibility tests. Some involve patient-related factors. Some involve pathogen-related factors. However one explanation is error arising from a deficiency in the antibiotic susceptibility test itself. That deficiency is that current routine antibiotic susceptibility tests do not detect the antibiotic-inactivating potential of some microorganisms. Some microorganisms produce enzymes that inactivate antibiotics. The best known enzymes of this type are the β-lactamases that certain bacteria produce to inactivate β-lactam antibiotics. Such enzymes, which are not reliably detected in routine antibiotic susceptibility tests, may cause sufficient antibiotic inactivation at the site of an infection in a patient to cause a treatment failure.
Plasmid-mediated AmpC β-lactamases were first reported in the 1980's. Bauernfeind, A., Y. Chong, and S. Schweighart 1989. Extended broad-spectrum β-lactamase in Klebsiella pneumoniae including resistance to cephamycins. Infection. 17:316-321. They arose as a consequence of the transfer of chromosomal genes for inducible AmpC β-lactamases onto plasmids. These enzymes have been reported in isolates of E. coli, K. pneumoniae, K. oxytoca, Salmonella spp., Citrobacter freundii, Enterobacter aerogenes, and Proteus mirabilis. The genes are typically encoded on large plasmids containing other antibiotic encoding resistance genes, leaving few therapeutic options. Although it has been over ten years since plasmid-mediated AmpC β-lactamases were discovered, most clinical labs and physicians remain unaware of their clinical importance. As a result, plasmid-mediated AmpC β-lactamases often go undetected. Without detection, correct therapy for infected patients may not be given.
Currently there are no NCCLS recommendations for detection of plasmid-mediated AmpC β-lactamases. The three-dimensional test has been used to detect these enzymes. However, this test is technically demanding and this has precluded its widespread adoption. Multiplex PCR is also currently available as a research tool for detection of plasmid-mediated AmpC β-lactamases, but is not yet available as a routine test for clinical labs. Perez-Perez F J, Hanson N D 2002. Detection of plasmid-mediated AmpC beta-lactamase genes in clinical isolates by using multiplex PCR. J Clin Microbiol. 40(6):2153-62.
The currently used antibiotic susceptibility tests, which measure only the antimicrobial activity of antibiotics and not the ability of the microorganisms to cause antibiotic inactivation, fail to take into account this very important determinant of the outcome of therapy. This deficiency in the tests places clinicians at a disadvantage in selecting the most appropriate antibiotics for their infected patients.
In summary, there is a need for antibiotic susceptibility tests that provide clinicians with information about both the antimicrobial activity of antibiotics and, additionally, the ability of microorganisms to inactivate antibiotics. Such tests should improve the quality of therapeutic decision-making by clinicians when selecting antibiotic therapies for patients with infections.
Antibiotic susceptibilities are determined routinely by either disk diffusion or antibiotic dilution methods, or by methods that are derivatives of these two methods. In disk diffusion methods [for example see Bauer, A. W. et al., American Journal of Clinical Pathology. 45:493-496 (1966); Bell, S. M., Pathology. 7:Suppl 1-48 (1975); Stokes, E. J., et al., Association of Clinical Pathologists Broadsheet, No. 55 (revised) (1972)] a standard quantity of the causative microorganism is uniformly spread over the surface of an appropriate culture medium (hereafter referred to as agar). Then several filter paper disks impregnated with specific amounts of selected antibiotics are placed on the agar surface. The agar is then incubated for an appropriate period at an appropriate temperature. During incubation the antibiotics diffuse out of the disks into the agar and the microorganism grows on the surface of the agar, except in the areas where antibiotics inhibit its growth. Inhibition of growth is detected as clear zones of no growth (inhibition zones) on the agar around the antibiotic disks. The sizes of the inhibition zones are measured and compared to established interpretive criteria to determine the microorganism's susceptibility or resistance to antibiotics.
Dilution methods test antibiotic susceptibility on either solid agar or in liquid broth. [National Committee for Clinical Laboratory Standards 1997. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. Approved standard M7-A4. National Committee for Clinical Laboratory Standards, Villanova, Pa.] In the dilution method a constant quantity of microbial inoculum is introduced into a series of tubes or wells of broth (or onto one or more agar plates) containing varying concentrations of antibiotic. After incubation for an appropriate period the broth (or agar) tests are inspected and the lowest concentration of antibiotic that prevents detectable growth of the microorganism recorded. This concentration is the minimum inhibitory concentration (MIC) of the antibiotic.
Both disk diffusion and dilution methods are generally deficient in that they do not yield information about the ability of microorganisms to inactivate antibiotics.
Various techniques for detecting antibiotic-inactivating enzymes of bacteria have been reported in the scientific literature. Some techniques are used to detect the activity of specific antibiotic-inactivating enzymes (e.g. β-lactamases), while others are non-specific and detect enzymes that inactivate more than one class of antibiotics. The following are exemplary tests used to detect antibiotic-inactivating enzymes of bacteria.
Specific tests for the detection of chloramphenicol acetyltransferase are reviewed in Chauchereau, A., et al., Analytical Biochemistry. 188:310-316 (1990). These are complex tests to detect enzymatic inactivation of chloramphenicol and require special instruments capable of measuring the absorbance of light at specific wavelengths or of measuring radioactivity. Such tests are not antibiotic susceptibility tests and their complexity is such that they are unsuitable for routine clinical microbiology laboratories.
The production of β-lactamases by Staphylococcus aureus is inferred by the production of a distinctive heaped-up margin of the inhibition zone around a penicillin antibiotic disk. Gill, V. J., et al., J. Clin. Microbiol. 14:437-40 (1981). Beta-lactamase production by many types of bacteria can also be detected chemically by testing the bacteria with an indicator substance such as nitrocefin. Oberhofer, T. R., et al., J. Clin. Microbiol. 15:196-9 (1982); O'Callaghan, C. H., et al., A.A.C. 1:283-288 (1972). These tests are reliable indicators only of β-lactamase-determined resistance of Staphylococcus aureus, Staphylococcus epidermidis, Moraxella catarrhalis, Neisseria and Haemophilus species to certain types of penicillin antibiotics. They do not predict the potential for any other bacteria to resist these penicillins, and they do not predict the potential for any bacteria to be resistant to any of the other classes of β-lactam antibiotics, such as cephalosporins, cephamycins, monobactams, monocarbams, penems or carbapenems. In short, these are useful tests of limited scope. For tests of β-lactam antibiotics, a more comprehensive test is needed to detect the activities of all β-lactamases against all β-lactam antibiotics.
Disk diffusion tests can be modified by a pre-incubation procedure to determine the ability of β-lactamases from Staphylococcus aureus to inactivate β-lactam antibiotics. Lacey, R. W., and A. Stokes, J Clin Pathol. 30:35-9 (1977). This procedure results in smaller inhibition zones than those for which the interpretive criteria of the tests were calibrated. The preincubation procedure thereby invalidates the interpretive tables that are necessary to determine antibiotic susceptibility or resistance. This is a serious deficiency because it would be unethical to base therapy on this procedure which lacks validated interpretive criteria.
The clover leaf test [Andremont, A., et al., Proceedings Reunion Interdisciplinaire de Chimiotherapie Antiinfectieuse. Societe francaise de Microbiologie, Paris:50 (1982); Kjellander, J., et al., Acta Pathologica Microbiologica Scandinavica. 61:494 (1964)] is a special technique used to detect β-lactamases and is also claimed to detect two other types of antibiotic-inactivating enzymes, chloramphenicol acetyltransferase and erythromycin esterase. This test is not an antibiotic susceptibility test and must be set up as an additional procedure. This is inconvenient and therefore a disadvantage. Furthermore there are doubts about the validity of results obtained with this procedure. Jorgensen, P. E., Chemotherapy. 31:95-101 (1985); Reig, M., et al., European journal of Clinical Microbiology. 3:561-562 (1984).
The cefoxitin induction test [Sanders, C. C., and W. E. Sanders, Jr., A.A.C. 15:792-797 (1979)] is a special procedure for detecting a particular type of bacterial β-lactamase, the inducible AmpC β-lactamase of Bush Group 1. Bush, K., et al., A.A.C. 39:1211-1233 (1995). This test does not detect all types of β-lactamases, and like the clover leaf test it is a special procedure that can be used to supplement antibiotic susceptibility tests. It is not, in itself, an antibiotic susceptibility test.
The double disk potentiation test (and its derivatives) involves strategically placing an amoxicillin/clavulanate or ticarcillin/clavulanate disk 20 to 30 mm from disks containing cefotaxime, ceftriaxone, ceftizoxime, ceftazidime, cefepime or aztreonam on an agar plate. It is therefore possible to determine if a strain of Enterobacteriaceae produces a special type of β-lactamase known as an extended-spectrum β-lactamase. Brun-Buisson, C., et al., Lancet. ii:302-306 (1987). The test is based on the ability of the β-lactamase inhibitor, clavulanate, to inhibit the extended-spectrum β-lactamase and prevent it from inactivating the cephalosporin or aztreonam antibiotics in the test. This is a special procedure, not a routine antibiotic susceptibility test, and detects only certain types of β-lactamases. It is therefore inconvenient and limited in scope.
A variety of disk and dilution tests have been derived from the principle of the double disk test. Brown, D. F., et al., J. Antimicrob. Chemother. 46:327-8 (2000); Cornican, M. G., et al., JCM. 34:1880-1884 (1996); Ho, P. L., et al., JAC. 42:49-54 (1998); Moland, E. S., et al., JCM. 36:2575-9 (1998); Sanders, C. C., et al., JCM. 34:2997-3001 (1996); Schooneveldt, J. M., et al., Pathology. 30:164-8 (1998); Thomson, K. S., et al., Antimicrob. Agents Chemother. 43:1393-400 (1999). That is, they use the ability of a β-lactamase inhibitor to inhibit an extended-spectrum β-lactamase to detect this type of β-lactamase.
The 3-dimensional test [Thomson, K. S., et al., J. Antimicrob. Chemother. 13:45-54 (1984); Thomson, K. S., and C. C. Sanders, A.A.C. 36:1877-1882 (1992); U.S. Pat. No. 5,466,583] is an approach that partially fulfills the need for improved antibiotic susceptibility testing.
In performing the direct form of the 3-dimensional test a standard quantity of the causative microorganism is uniformly spread over the surface of an agar plate in the usual manner for performing a disk diffusion test. However, before placing the antibiotic disks onto the surface of the agar, the 3-dimensional inoculation is performed. This is effected by using a sterile scalpel to cut a slit in the agar 3 mm to one side of where the antibiotic disks will be placed. A dense liquid inoculum of the test microorganism is then dispensed into the slit, the antibiotic disks are placed on the agar 3 mm from the slit, and the test is incubated.
After incubation the inhibition zones are measured by standard procedures to determine the susceptibility or resistance of the microorganism to the test antibiotics according to the interpretive criteria of the disk diffusion test. However, in addition to this, enzymatic inactivation of the antibiotics can be detected by inspecting the intersections of the 3-dimensional inoculum with the margins of the inhibition zones. Antibiotic inactivation results in a distortion or discontinuity in the usually circular inhibition zone. (These distortions or discontinuities are hereafter referred to as “3-dimensional effects”.)
The 3-dimensional test thus allows the laboratory to report to the clinician not only the susceptibility or resistance of a microorganism to antibiotics, but also the ability of the microorganism to inactivate the antibiotics. As a hypothetical example, whereas a conventional antibiotic susceptibility test might indicate that a microorganism was susceptible to the two antibiotics, cefaclor and cefoxitin, the 3-dimensional test might provide additional information to show that the microorganism produced an enzyme capable of inactivating cefaclor but not cefoxitin. Thus, although the conventional test indicated that both antibiotics appeared to be equally efficacious, it would appear, from the additional information provided by the 3-dimensional test, that only cefoxitin might not be inactivated in the patient and therefore would constitute a more effective treatment than cefaclor. In this example, the information provided by the 3-dimensional test could assist a clinician to make a better choice of therapy.
In addition to the direct form of the 3-dimensional test, the indirect form is used for testing microorganisms when inhibition zones are small or absent, or as a research or diagnostic method. The indirect test is performed by inoculating the surface of the agar with a fully susceptible assay strain such as Escherichia coli ATCC 25922. After this, the 3-dimensional slit is cut in the agar and inoculated with a suspension of the test microorganism. Although the indirect test precludes the simultaneous determination of antibiotic susceptibilities, it permits investigation of the antibiotic inactivating enzymes of microorganisms for which the inhibition zones are too small to yield 3-dimensional results when the test is performed in the direct form of the test.
There are several problems with the 3-dimensional test. These problems include the following:                a. The procedure for making the slit in the agar for the 3-dimensional test is inconvenient and technically difficult to perform correctly.        b. Making the slit is potentially dangerous to laboratory staff because a scalpel blade contaminated with pathogenic bacteria is an infection hazard.        c. It is also technically difficult to accurately deliver the liquid 3-dimensional inoculum into the slit without overfilling the slit and possibly invalidating the test.        
A variety of chemicals have been reported to disrupt or permeabilize microbial membranes, thereby increasing their permeability and causing loss of cellular contents. These include certain antibiotics, detergents, chelating agents, polycations, hydrophobic dyes, and enzymes. Nikaido, H., and M. Vaara, Microbiol Rev. 49:1-32 (1985); Piers, K. L., et al., Antimicrob. Agents Chemother. 38:2311-2316 (1994). These chemicals are often bacteriostatic and/or bacteriocidal in normal use.